Dynamic tail risk contagion in multi-asset markets: A DCC-MGH framework with CoEVaR analysis during systemic crises

• A DCC-MGH-CoEVaR framework captures multi-market tail risk spillovers. • Crude oil acts as a persistent net transmitter during crises. • Gold shifts from absorber in COVID-19 to redistributor in the war. • Bonds buffer risks while stocks amplify spillovers across markets. To investigate the risk spillover effects between crude oil and financial markets (including stocks, bonds, and gold).This study introduces a novel Dynamic Conditional Correlation-Multivariate generalized hyperbolic distribution(DCC-MGH) model combined with the Conditional Entropic Value-at-Risk(CoEVaR) approach to address the limitations of traditional models regarding distributional assumptions and insensitivity to extreme losses. The framework aims to accurately capture dynamic correlations and asymmetric tail risks across multiple markets. The findings show that:(1) crude oil exhibits persistent risk spillover characteristics throughout the entire sample period;(2) gold demonstrates particularly prominent dynamic safe-haven properties during crises;(3) bonds possess stable hedging attributes while stocks serve as risk transmission hubs.These results provide new theoretical foundations for enhancing global financial security and optimizing asset allocation strategies.They also offer policy guidance for cross-market risk management under unconventional shocks.